Robot

Transfer of learning

Inductive transfer

Computer security

Economics

Set (abstract data type)

Computer science

Robot learning

Code (set theory)

Language model

Natural language processing

Task (project management)

Management

Mobile robot

Scratch

Artificial intelligence

Programming language

Machine learning

Key (lock)

Inductive transfer learning has greatly impacted computer vision, but existing approaches in NLP still require task-specific modifications and training from scratch. We propose Universal Language Model Fine-tuning (ULMFiT), an effective transfer learning method that can be applied to any task in NLP, and introduce techniques that are key for fine-tuning a language model. Our method significantly outperforms the state-of-the-art on six text classification tasks, reducing the error by 18-24% on the majority of datasets. Furthermore, with only 100 labeled examples, it matches the performance of training from scratch on 100 times more data. We open-source our pretrained models and code.

Universal Language Model Fine-tuning for Text Classification